issue
Pass
Audited by Gen Agent Trust Hub on Jul 15, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to process untrusted user data to create and update GitHub issues, which presents a surface for indirect prompt injection.
- Ingestion points: User-provided issue titles, bodies, and comments.
- Boundary markers: The skill instructs the agent to avoid passing multiline Markdown inline through shell arguments, instead recommending the use of temporary files and the
--body-fileflag. - Capability inventory: The skill leverages
gh issue create,gh issue edit, andgh issue commentto perform write operations to GitHub. - Sanitization: It promotes the use of quoted heredocs (
<<'EOF') or thewritetool to prevent shell expansion or interpolation of the data being written to temporary files. - [COMMAND_EXECUTION]: The skill uses the GitHub CLI (
gh) to interact with the GitHub API. This is the intended functionality of the skill and follows standard development workflows.
Audit Metadata